2007-cje[1]

7/30/2019 2007-cje[1]

1/23

The ontology of economic power incapitalism: mainstream economicsand Marx

Giulio Palermo*

Mainstream economics conceives power to be incompatible with perfect competi-

tion. This conception, I argue, derives from its deductivist method and the empiricalrealist ontology that it presupposes. Following Marx, I show that capitalismconstantly reproduces asymmetrical constraints on classes of individuals, indepen-dent of the market form. My ontological argument is rooted in the philosophy ofcritical realism. My conclusion is that the dichotomy powercompetition isontologically untenable. If capitalist relations necessarily involve power, it is notsimply because neoclassical competition does not exist in reality, as radicalmainstream economists suggest, but rather because capitalism altogether is a systemof power.

Key words: Power, Marxism, Critical realism, Ontology, MethodologyJEL classifications: B25, B41, D40, P16.

1. Introduction

This paper brings together two notions that are not usually considered together: power and

ontology. Indeed, it is an ontological enquiry into the nature of power under capitalism

and, as a sub-text, the inability of mainstream economic theory to understand power

because of its mistaken ontological presuppositions.

In general terms, power is the faculty to do or to not do something, which I refer to as

power to act (PTA). In interpersonal relations, power is the capacity to influence the

action of others, which I refer to as power over somebody (POS). Whilst both aspects are

important, mainstream economics cannot explain PTA and its relations with POS becauseof its ontology, and so operates entirely with an analysis restricted to POS. This manifests

itself in the literature, via the twofold assumption that the distribution of PTA in society is

given and that, in a system of perfect (Walrasian) competition, there is no power in social

relations. This is because, in perfectly competitive markets, individuals with their inborn

endowments can choose among a number of equivalent interpersonal relations. Therefore,

if an individual tries to impose his/her will upon another, the latter can simply choose to

Manuscript received 15 January 2002; final version received 21 July 2006.Address for cor respondence: Dept of Economics, University of Brescia, via S. Faustino 74/B, 25122 Brescia,

Italy; e-mail: [email protected]

* University of Brescia. A version of this paper was first presented at the EAEPE annual conference, Sienna,Italy, 2001. The author is grateful to Steve Fleetwood and to an anonymous referee of this journal for theircomments, critical remarks and suggestions.

Cambridge Journal of Economics 2007, 1 of 23doi:10.1093/cje/bel036

The Author 2007. Published by Oxford University Press on behalf of the Cambridge Political Economy Society.

All rights reserved.

Cambridge Journal of Economics Advance Access published January 10, 2007

7/30/2019 2007-cje[1]

2/23

stop interacting with him/her with no loss, independently of their respective PTAs. In

such a context, no one can actually exercise power over anyone else.

From a non-mainstream perspective, such as Marxist (as I shall elaborate here) or in-

deed a post-Keynesian or perhaps even Institutionalist one, capitalist relations are

recognised as, by their nature, power relations. Under capitalism, PTA is asymmetricallydistributed in society, and this asymmetry (which is necessary to the day-to-day operation

of capitalism) has nothing to do with the particular competitive or non-competitive form of

the market. And it is indeed this asymmetrical distribution of PTA at a social level which is

the cause of relations of POS at an interpersonal level.

How is it possible that Marxist (or other) perspectives can grasp this form of power,

whereas the mainstream perspective cannot? The answer lies not strictly at the level of

theory, but at that of meta-theory. It is Marxisms meta-theoretical, or more specifically,

ontological, presuppositions that discourage the development of theory in terms of isolated

individuals, and encourage the development of theory in terms of the relations between

agents and the social structures they interact with. And unsurprisingly, some of these social

structures are the vehicle through which power is asserted.

Mainstream economics presupposes an empirical realist (ER) ontology wherein,

essentially, what is empirically observed is treated as synonymous with what is: it is a kind

of what you see is what you get ontology. Marxist economics, by contrast, presupposes

a structured or depth ontology wherein, underlying the empirically observed and the actual

occurrences, are social structures that govern the empirical and the actual. This ontology, I

argue, is compatible with the philosophical perspective of critical realism (CR). Put

simply, for those like critical realists and Marxists committed to a structured ontology,

essence underlies appearance, whereas for mainstream economists, committed to an ER

ontology, appearance is all there is. These two different ontologies, then, motivate and

sustain two different notions of power.This paper starts with ontology. It elaborates upon the structured ontology developed

within CR and the ER alternative before going on to discuss the individualist approach to

power (developed by mainstream economics, but also accepted in some heterodox

accounts) encouraged by ER. Bearing ontology in mind, I consider the relations between

the different forms of power. As well as the empirical notions of PTA and POS, I shall

discuss non-observable entities such as constraining structures, constraining mecha-

nisms and system of power. This abstract ontological framework is used in the following

section to analyse the concrete forms of economic power in capitalism. The structure of this

section follows the preceding one. Here, however, PTA, POS and the other power-related

concepts are discussed in the context of capitalist relations as purchasing power, marketpower, authority, etc. This analysis relies heavily on Marxs understanding of the working

and reproduction of capitalism. My thesis is that capitalism is in its essence a system of

power, with particular empirical manifestations of an underlying non-observable power

structure.

2. Critical realism

The CR movement begins in the philosophical camp with the work of Bhaskar in the

1970s, and is developed in the 1980s thanks to the contributions of Benton, Collier and

Outhwaite (cf., Norris, 1999). Today, CR is a movement encompassing different

disciplines, including economics, sociology, biology and physics. The growing impact ofthis movement is witnessed by the number of associations, forums and workshops fostering

2 of 23 G. Palermo

7/30/2019 2007-cje[1]

3/23

the discussion and development of CR and by the increasing penetration in academic

journals of themes related to CR. Since 1998, the International Association for Critical

Realism has also published a specialised journal on CRthe Journal of Critical Realism. In

the context of philosophy and economics, the works of Archer (1995), Bhaskar (1978,

1979, 1986, 1993), Collier (1994), Lawson (1997, 2003) and Outhwaite (1987) constitutesome of the main theoretical references. Some essential readings on CR are collected in

Archer et al. (1998). Fleetwood (1999) is a collection of contributions to the debate about

CR in economics. The development of a CR perspective in economics owes much to the

Cambridge school, strongly influenced by Tony Lawson (some of its contributions have

appeared in this journal and in the Review of Social Economy).

According to CR, mainstream economics can be characterised by its deductivist

methodology, based on closed system modelling. Deductivism, like any other method,

presupposes an ontology, in the sense that the nature of reality must be supposed to be such

that it can be investigated using the deductivist method. Although mainstream economics

rarely develops ontological arguments, its implicit ontology consists of atomistic, empirical

events. More precisely, reality is supposed to be constituted by two domains, the actual

and the empirical. The former consists of events and states of affairs, the latter of human

experiences of them. These two domains are supposed to be fused, so that actual events are

presumed to coexist with their empirical perceptions. Reality is thus identified with what is

perceived, or, at least, what is perceivable under certain conditions, and the explanation of

an event is provided by looking at other (perceivable) events, since nothing else exists in

this ontology. If knowledge is, allegedly, gained through observing and recording these

events empirically (as data), then generalised and/or scientific knowledge can only be

gained if these events produce some kind of stable patternif they simply produces

a totally random flux of events, on this understanding, there could not be access to

knowledge. In fact, the pattern these events produce has to be one of event regularity orconstancy. Mainstream economists, typically, generate these event regularities by assump-

tion. The result is an artificially closed system. The snag is that socio-economic systems are

almost always open systems, i.e., systems wherein event regularities are not ubiquitous.

By contrast, according to CR, the world is composed not only of events and states of

affairs and our experiences or impressions of them, but also of underlying structures,

powers, mechanisms and tendencies, which exist, whether or not detected, and govern or

facilitate actual events (Lawson, 1997, p. 21). The CR ontology is stratified and consists of

three domains, the empirical, the actual and the real (also called the deep). These

three domains are presumed to be out of phase with one another. This means that it is not

possible to establish a one-to-one relationship between experiences and events or betweenevents and mechanisms. This ontology encourages a different methodology from

deductivism. In contrast to the deductivist mode of inference, observable events are not

explained in terms of other observable events, but in terms of underlying structures, their

causal powers, the mechanisms through which they operate and the tendencies that they

generate. This move from phenomenal reality to the deep domain is a mode of inference

called retroduction. Structures and mechanisms, however, do not necessarily act in

isolation; on the contrary, in general, an event or state of affairs is governed by the

interaction of different structures and mechanisms. The scientific problem is thus to

identify the structures and mechanisms that govern phenomenal reality and to explain the

forms of their interaction.

These different ontological premises lead to radically different conceptions of power.According to CR, power is a notion that refers to entities of the deep domain. By contrast,

The ontology of economic power in capitalism 3 of 23

7/30/2019 2007-cje[1]

4/23

in ER inspired theories, the notion of power refers to particular empirical phenomena.

From a CR perspective, the problem is thus to explain the relation between the empirical

forms of power and the structures and mechanisms that govern them. In the study of power

relations, however, the implicit or explicit acceptance of ER by academic orthodoxy has

encouraged a completely different research project, based on the deductivist method. Infact, the most common way to study power relations has been to assume atomistic

individuals as the sole explanatory units of eventual relations of POS (methodological

individualism). But in order to close the system and make deductivism applicable,

atomistic individuals also had to be deprived of any real agency. Methodological in-

dividualism has thus been coupled with the assumption of rational economic men. Only

such a man is analytically empty, analytically atomistic, so that there is no intrinsic

properties that can change, thereby generating different responses to the same stimuli at

different times. Only an atomistically conceived homo economicus responds identically, and

predictably every single time he is faced with the same set of conditions.1 It is within this

methodological framework, emanating from the ER ontology, that orthodox social

scientists have approached the study of power.

3. The individualist approach

The subject of power has been investigated mainly within sociology and political science.

The individualist approach has its roots in Weber, who defined power as the probability

that one actor within a social relationship will be in a position to carry out his own will

despite resistance (Weber, 1968, vol. 1, p. 53).2 This definition has been applied mainly to

the study of situations in which an actor is able to get his/her way in social decisions when

others are openly opposed. According to Dahl (1961), the exercise of power presupposes

that two or more groups have conflicting preferences and that they manifest this explicitly.Lukes (1974), however, argues that this is a restrictive conception of power that can be

called one-dimensional, and identifies two other views of power: the two-dimensional,

developed by Bachrach and Baratz, and his own three-dimensional view.3

Bachrach and Baratz (1970) introduce a second dimension of power that consists in the

ability to condition the issues that are the object of collective decision. If A manages to

confine the scope of decision-making to particular issues and prevents B from bringing to

the fore issues that might be detrimental to his/her own preferences, A is actually exercising

power over B. This two-dimensional view, however, is still inadequate for Lukes. The

political agenda, he argues, is not necessarily controlled by the intentional action of

particular individuals; it depends also on collective action and on the form of organisationof the system (systemic effect). Moreover, the notion of power should not be restricted to

observable conflicts, because power over an individual may also be exercised by in-

fluencing, shaping or determining his/her very wants. Within this framework, even the

absence of grievance does not imply genuine consensus, since those who are subject to

power might not be able to express their realinterests and might even be unaware of them.

And most importantly, there might exist forms of coercion on individual choice that

1 This explains why Austrian economics, the champion of methodological individualism, but also a severecritic of neoclassical homo economicus, has never really participated to the debate on power within mainstreameconomics.

2 The English translation can be misleading, for the German term chance (translated as probability)

also means opportunity.3 After the criticisms of Bachrach and Baratz against the work of Dahl, the one-dimensional view has beendefended by Polsby (1963) and Wolfinger (1971).

4 of 23 G. Palermo

7/30/2019 2007-cje[1]

5/23

depend on the overall structure of the decision-making system and that cannot be ascribed

to the action of any single individual.

Besides these conceptions of power as an interpersonal relation, Lukes (1974, p. 31)

criticises the conceptions elaborated by Parsons (1957, 1963A, 1963B) and Arendt

(1970), by noticing that they focus on the locution power to, ignoring power over.Thus power indicates a capacity, a facility, an ability, not a relationship. Ac-

cordingly, the conflictual aspect of powerthe fact that it is exercised over people

disappears altogether from view. Lukes also maintains that everything that can be said by

means of the notion of power to can be said with greater clarity by means of his own

conceptual scheme. Although my analysis of economic power is an attempt to develop the

three-dimensional view of Lukes, I do not think that the point is to establish whether the

notion of POS is more or less general than that of PTA. The issue is rather to clarify the

relations between them and, above all, the way both of them are governed by non-

observable structures and mechanisms. In fact, I believe that this is the only way to study

systemic effects, which Lukes himself regards as an essential aspect of his three-

dimensional view.1

In the economics literature, the debate on power has developed mainly as a sub-problem

of explaining the nature of the firm. Coase (1937) formulated this problem by asking the

questions: Why do hierarchies exist? Where do power relations within the firm come

from?2 Of course, these questions can be approached in many ways. The acceptance of

deductivism, however, and, in particular, of methodological individualism, poses serious

restrictions that make the one-dimensional view largely dominant even within important

parts of economic heterodoxy. Meta-theoretically speaking, even those parts of Marxism

committed to deductivism and ERsuch as rational choice Marxism and parts of the

radical schooldevelop in fact the same one-dimensional view and are indistinguishable

from mainstream economics.This one-dimensional view includes traditional neoclassical economics, with its analysis

of market power, the contractual approach (Alchian and Demsetz, 1972; Alchian and

Woodward, 1987; Jensen and Meckling, 1976; Cheung, 1983, 1987A, 1987B, 1992), the

transaction cost economics (Coase, 1937; Williamson, 1975, 1985, 1995, 1996A, 1996B;

Williamson and Ouchi, 1983; Simon, 1951, 1991), the property right approach (Hart,

1995, 1996; Hart and Moore, 1990), and those parts of the institutional perspective

(Goldberg, 1976, 1980) and of the radical school (Bowles and Gintis, 1988, 1993A,

1993B, 1993C, 1993D, 1994; Bowles, 1985; Gintis, 1976, 1989; Gintis and Ishikawa,

1981; Putterman, 1982, 1984, 1993, 1995; Dow and Putterman, 1999, 2000; Screpanti,

2001) that accept methodological individualism even if only as a theoretical challenge.

3

1 The issue of power in political science is often associated with the work of Foucault. In his analysis,however, Foucault addresses various questions on different aspects of power, but, as he himself claims, he inno way construct[s] a theory of power (Foucault, 1990, p. 39). In his conception, power exists in a purelynominal sense rather than in any substantive sense: power is the name one attributes to a complex strategicalrelationship in a particular society (Foucault, 1980A, p. 93). As Cousins and Hussain (1984) notice,Foucaults major interest is not in what is power, but in how it is exercised. As he says: Power is exercised ratherthan possessed (Foucault, 1977, p. 26); it only exists in action (Foucault, 1980B, p. 89). He also has nointerest in the quantification of power and disputes whether power comparability in any quantifiable formexists.

2 An equivalent formulation of the same question is that of Samuelson (1957) who asks why, in perfectlycompetitive markets, capital usually hires labour and not vice versa.

3

In Palermo (2000), I argue that new institutional economics fails both its attempts (1) to characterisetheoretically the capitalist firm and (2) to analyse the power relations of capitalism. In Ankarloo and Palermo(2004), we focus the critique on Williamsons transaction costs economics.


7/30/2019 2007-cje[1]

6/23

If I assemble theories that are often considered as competing with one another, it is

because their individualist methodology and insistence on observable conflicts presuppose

the same ER ontology. Ontologically, these theories assume that reality coincides with what

is empirically detectable. Methodologically, they seek (what they consider to be)

explanations by reducing empirical phenomena to the smallest units of the system, whichare assumed to be independent of one another and of the system of which they are part.

These units, as already noticed, are rational economic agents. Although this approach

focuses on the relationship between the capitalist and the worker, these two figures are not

bona fide social entities. In this ontology, capitalists and workers are simply individuals.

Therefore, their eventual power relation is not found in the capitallabour relation, but in

the innate qualities of individuals.

New Institutionalism, for instance, explains the origin of hierarchies in capitalism by

assuming that they emerged spontaneously from the interaction of imperfect individuals in

an imperfect context. Williamson (1975, p. 21) starts with the assumption that in the

beginning there were markets and, through successive exercises in comparative statics,

deduces the emergence of hierarchies as solutions to market failures. His explanation is

based on the assumption that individuals have heterogeneous natural endowments, such

as unequally distributed administrative talent, oratorical gifts, information processing

and decision making skills (Williamson, 1975, pp. 4752). As Pitelis (1991, p. 12) has

shown, however, heterogeneity in natural individual endowments alone is not sufficient to

make hierarchies superior to markets. It is necessary that the context in which individuals

interact be imperfect as well. For this reason, Williamson introduces uncertainty and

imperfect information as essential features of his market and hierarchies framework.

The assumption of an everlasting human nature makes Williamsons historical analysis

peculiar: the statement that capitalist hierarchies originated without coercion is not

documented historically, but argued deductively. The problem for Williamson is not toinvestigate what has effectively taken place, but to find the conditions that make one

institution superior to another, with the (unjustified) assumption that efficiency is always

automatically selected. In this conception, hierarchies can exist only where the conditions

for perfect competition do not hold.1

This conception is common to the whole of the individualist approach. In a perfectly

competitive context, interpersonal relations are assumed to involve no power, and the

cause of power relations is found in the empirical conditions that make perfect competition

impossible.2 In the words of two leading exponents of radical political economics, the

absence of power in the Walrasian model is based on the presumption that supply equals

demand in competitive equilibrium for, when markets clear, each agents transaction isequivalent to his or her next best alternative (Bowles and Gintis, 1994, p. 301). According

to the authors, the strength of two individuals who interact in the market depends on the

difference between their current relation and their next best alternative. If a capitalist pays

1 The individualist approach assumes that individuals are self-interested. However, self-interestedbehaviour is depicted as profit maximisation with regard to the capitalist and as shirking or cheatingwith regard to the worker. Notice also that competition and power are conceived as mutually exclusive andsymmetrical, but are analysed asymmetrically: power is to be explained; competition, by contrast, is naturaland everlasting and deserves no scientific explanation.

2 This conception of power can be extended to Austrian economics as well. As Young (1995) shows,although Austrians consider the market as an arena of freedom in which interaction is purely voluntary, their

conception of capitalism as essentially free from power relations presupposes atomism and competition. InPalermo (1998), I criticise the Austrian conception of the market process for its neglect of power relations inthe explanation of the convergence of individual plans.

6 of 23 G. Palermo

7/30/2019 2007-cje[1]

7/23

a wage that is higher than the perfectly competitive one, then he/she holds the worker in

check, and has power over him/her. But in the reign of perfect Walrasian competition, this

cannot occur.1

Perfect competition presupposes full rationality, perfect information, and absence of

radical uncertainty and historical time. Therefore, according to this approach, powerrelations can exist only in empirical contexts characterised by these sorts of imperfection. At

an ontological level, the actual-empirical domain is so divided into two distinct closed

systems: a system without any imperfection, in which interpersonal relations are governed

by perfect competition; and a system with imperfections, in which interpersonal relations

involve power. This distinction might appear purely theoretical. However, it is only when

such a distinction exists in reality that it is meaningful to develop different and

incompatible theories for both perfect and imperfect decision-making contexts: only in

this case can the theories developed in the two decision-making contexts be simultaneously

valid, each one in its subset of the actual-empirical domain. If, on the contrary, the perfect

decision-making context is a pure fiction, with no empirical counterpart, then, to be

rigorous, the whole set of theories built in it should be considered irrelevant in realist

terms.2

The extent of power relations in reality is thus a merely empirical question, depending on

the spread of imperfections in the actual-empirical domain. At one extreme, we find

Alchian and Demsetz, who assume markets and perfect competition even where they do not

exist, as in the case of intra-firm relationships. They write as follows:

It is common to see the firm characterized by the power to settle issues by fiat, by authority, or bydisciplinary action superior to that available in the conventional market. This is delusion . . . [Theemployer] can fire or sue, just as I can fire my grocer by stopping purchases from him or sue himfor delivering faulty products.3 (Alchian and Demsetz, 1972, p. 777)

At the other extreme, authors of radical inspiration see market imperfections as endemic

and maintain that power relations are ubiquitous in the real world. For instance, Screpanti

(2001, p. 145), after defining a complex context characterised by bounded rationality,

imperfect information, uncertainty and various externalities, affirms: Perfect and atomistic

competition cannot exist in this world, even as a limit caseI mean the neoclassical

competition that eliminates all inefficiencies and power hubris. But these apparently

opposite positions share the same conception, according to which the pervasiveness of

power relations in reality depends on the empirical diffusion of market imperfections, which

1 Bowles and Gintis distinguish between perfect competition and perfect Walrasian competition: theformer does not imply market clearing, the latter does (and coincides with what in the literature is generallycalled perfect competition). Ultimately, this is why they can claim that power relations exist even in the reignof perfect competition (with non-clearing markets). In this paper, I shall continue to adopt the definition ofperfect competition as implying market clearing.

2 To the extent that in fact the perfect decision-making context does not exist in reality, Bowles and Gintisare coherent in rejecting the orthodox Walrasian model. Less clear is why they choose precisely this abstractfiction as the starting point of their supposedly more realistic post-Walrasian approach.

3 Within new institutional economics, Williamson (1997, p. 23) explicitly criticises Alchian and Demsetz(1972) by affirming: Firms can and do exercise fiat that markets cannot, and characterises intra-firmrelations precisely on the asymmetry between employers and employees. Williamson, however, considers thenotion of power intrinsically problematic and analyses the institutions of capitalism under the assumptionthat they are Pareto-efficient outcomes of free voluntary exchanges. Fourie (1989, 1991, 1993) has argued

that this explanation of hierarchy in production rests on a theoretical confusion between production andexchange. Before him, Engels (1878, Part II, ch. 1) had noticed, Production may occur without exchange,but exchangeby the very fact that it is only an exchange of productscannot occur without production.


7/30/2019 2007-cje[1]

8/23

prevents markets from clearing and breaks the symmetry between the two sides of the

market.

Before developing an explicit ontology that explains the social nature of capitalist power

relations, let me briefly discuss a fundamental contradiction in the individualist

methodology and in the ahistorical conception of capitalism it presupposes. Boundedrationality, imperfect information, historical time, etc. are not specific to capitalism;

therefore, if we follow the deductivist logic of mainstream economics, power relations must

be a constant in any social system. But then we must conclude that power relations have

always existed, even before the historical development of market relations and economic

competition, although they have become visible only with the historical development of

capitalism (and the consequent possibility of conceiving a model of complete markets and

perfect competition). However, if pre-capitalist systems, with less developed or completely

absent market relations, were not regulated by economic competition, this clearly could

not have been due to market imperfections, but to lack of market relations. As I shall show, the

claim that power relations are ubiquitous in capitalism is correct, but for completely

different reasons. But before analysing the specific forms of economic power of capitalism,

I shall develop a general ontological framework of the relations between the different forms

of power.

4. The ontology of power

4.1 Power to act

The PTA is defined by the decision-making set of the agent, which exhaustively describes

his/her potential courses of action. Agents are not only individuals, but also collective

actors (a firm, a bank, a political party, a trade union . . . ). There is no reason to assume

that they know precisely and take into account rationally their own decision-making sets

and those of the others. Imperfect information and bounded rationality can lead to partial

and imperfect knowledge. Independent of this, however, the sets of actions that agents can

and cannot undertake existand do not depend on what agents know. Of course, the analysis

of the existing PTA is not sufficient to determine what agents will do, or, to put it

differently, the analysis of the existence of PTA cannot explain how agents exercise it: choices

do not depend only on constraints (or, better, on the perception and interpretation of

them), but on goals as well.

Operationally, at a given moment of time, a measure of the existing PTAs is defined by

the range of actions inscribed in each individual decision-making set. This measure is ex

ante in the sense that it refers to the means at ones disposal for the exercise of power, not tothe exercise of power itself: purchasing power is determined by wealth, not by actual

purchases; military power is determined by fire potential, not by the tons of bombs actually

dropped. When a power is exercised, its measurement is ex post. As concerns existence, this

measure is a mere imperfect proxy for the ex ante measure: purchases are correlated with

wealth, bombs with fire potentials, but there is no necessary causality between them.

Therefore, even remaining within the actual-empirical domain, empirical correlation is not

an adequate tool for the analysis of existence of power relations.

4.2 Power over somebody

A has power over B when the choice ofB depends on the specific action chosen byA withinhis/her own decision-making set. This can occur, first, when A is able to influence Bs

8 of 23 G. Palermo

7/30/2019 2007-cje[1]

9/23

goals, and, second, when A has some control ofBs decision-making set (in this case, POS

can be seen as the ability to change the PTA of other agents).1

POS and PTA are intimately linked, since the emergence of a relation of POS depends

on the distribution of PTAs in the system. In a system of complex interpersonal relations,

the decision-making sets and the goals of agents are necessarily interdependent. In thissense, relations of POS are only the tip of the iceberg that emerges from a sea of relations of

PTA. The strongest form of POS is the authority relation, in which A can order B to do

something and B must obey: in terms of constraints, As authority over B is expressed by

As ability to restrict Bs decision-making set to just one option.2

Operationally, an ex ante measure of the existing relations of POS must be defined along

at least two dimensions, an extensive one and an intensive one. The former is given by the

number of agents whose behaviour depends on the action undertaken by A. The latter is

given by the importance of the consequences that a particular decision of A produces on

each single individuals goals and constraints. These dimensions are conceptually

independent: on the one hand, it is possible that A has power over many agents but that

his/her power over each of them is weak; on the other hand, it is possible that A has power

over only an agent but that this power relation takes its strongest form, namely authority.

Ex post measures can be provided only when the constraints imposed by A on the

decision-making set of B are binding, given Bs goals. If, on the contrary, A modifies Bs

goals or constraints, but B does not change his/her choice, this relation of POS cannot be

detected by the empirical observation ofBs behaviour. This means that empirical analysis

can reveal only a sub-set of the existing relations of POS.

4.3 The constraining structure

The set of individual decision-making sets and their interdependences define a decision-making system. Within a decision-making system, the relations among individual

decision-making sets can take different forms. In some cases, individual decision-making

sets can simply be incommensurable, in the sense that A can undertake action x, but not

action y, whilst B can chose y but not x. In other cases, As decision-making set can be

a subset of Bs, which is to say that B has constraints that A has not. If this asymmetric

distribution of constraints is a necessary condition for the working of the system, then the

structure of the decision-making system is essentially a constraining structure. With this

term, I stress the asymmetric constraints that the structure of the decision-making system

1 According to Marxs historical materialism, individuals are free to act within the existing social context,

but are conditioned in their very consciousness by the mode of production of their material life (Marx, 1859,Preface; Marx and Engels, 1964, Part 1). This means that individual goals, like individual constraints, areinfluenced by the form of social interaction, independent of the existence of direct relations of POS. Withinheterodox economics, the institutional school in particular has studied the mechanisms through which powerrelations concur to shape individuals goals. Bartlett (1989) discusses how the values to be pursued, theauthority to be accepted, the rights to be respected, and the ideology to be internalised are all part ofendogenous social interaction. Galbraith (1983) analyses the forms and sources of power and the relationsbetween them. Dugger (1984, 1989) focuses on the process through which corporations impose their valuesand culture in society. Outside of economic debates, this issue is illustrated with rich empirical evidence byKlein (2000).

2 This definition of authority is inspired by the notion elaborated by Marx and Engels: the imposition ofthe will of another upon ours (Engels, 1872). It is also compatible with the definition of mainstreameconomics, although, within this literature, the notion of authority is also referred to as hierarchy,verticality or command. In political writings, the term authority is sometimes used to describe situations in

which B complies because he recognises that As command is reasonable in terms of his own values(Bachrach and Baratz, 1970, p. 34). This latter meaning takes inspiration from Webers (1968) conception oflegitimate authority that presupposes a consent to obey commands.


7/30/2019 2007-cje[1]

10/23

poses for different agents. However, this structure can also be understood as enabling

individual behaviours that could not be undertaken without its existence. These are just the

two sides of the same coin, observed from the angles of asymmetric constraints or

asymmetric PTAs.

4.4 Constraining mechanisms

When the constraining structure remains relatively stable in time, the problem arises of

discovering the mechanisms that regulate its reproduction and evolution. I shall call these

mechanisms constraining mechanisms, given their effects in the reproduction of the

constraining structure of the decision-making system.

4.5 Systems of power

When we discuss a particular decision-making system as a system of power, we do not

mean simply that the PTAs (and the POSs that they eventually produce) are precisely

defined. We mean something more, namely that a constraining structure and some

constraining mechanisms exist. This is what makes a decision-making system a system of

power.

As decision-making sets and goals can be influenced by completely different factors: in

one case, they can be (partly) controlled by B; in another case, they can be influenced by

the tendencies caused by the constraining mechanisms that govern the evolution of the

whole system of constraints, without anybody directly having power over him/her.

However, if the result is the same decision-making set and the same set of goals, then

the difference between the two cases is only formal, not substantial. This suggests that the

analysis of power cannot be reduced to the study of POS, for if two agents are part of

a system of power, their asymmetric PTAs can be systematically reproduced by impersonal

structures and mechanisms without necessarily giving rise to relations of POS.It is now time to move from a discussion of systems of power in general and consider the

case of a specific system, namely capitalism.

5. Capitalism as a system of power in Marxian political economy

Marx and Engels are, in my view, the authors who have best investigated capitalism as

a system of power. The relationship between Marxism and CR is not clear cut. Today, there

is a growing literature on this subject, which shows the coexistence of different, and at least

partly incompatible, viewpoints. A first, encouraging exploration of this relationship is

developed by the founder of CR himself, Bhaskar (1991, p. 143), who affirms that Marxswork at its best illustrates critical realism; and critical realism is the absent methodological

fulcrum of Marxs work. The compatibility between Marxs analysis and CR has been

defended at different levels, ranging from a simple formal coherence to a deep meta-

theoretical unity. Ehrbar (1998) goes so far as to say that Marx was a critical realist, long

before critical realism was born, and proposes a sentence-by-sentence translation of

Marxs Capitalinto CR terms (Ehrbar, 2001). The thesis of a mutually helpful relationship

is developed, in particular, by Collier (1979, 1989), Creaven (2001, 2003), Fleetwood

(2001) and Joseph (2001). On the opposite side, some overt criticisms about the

compatibility between Marxism and CR are developed by Gunn (1989), Magill (1994)

and Roberts (1999).

The main points of contrast and convergence between Marxism and CR are discussed ina book edited by Brown et al.(2001). In their introductory chapter, the editors explore

10 of 23 G. Palermo

7/30/2019 2007-cje[1]

11/23

three different viewpoints. According to Fleetwood, the nature of the relationship between

CR and Marxism is philosophical. In his view, CR can supply an all-encompassing

philosophy of science that is missing in Marxism and that allows us to place some Marxian

notions on a more secure philosophical footing. By contrast, Roberts argues that Marxism

is in no need of the philosophical services of CR and suggests that a more suitable wayfor Marxism to proceed would be to develop the theoretical categories of historical

materialism, rather than incorporate some concepts and categories of CR, which, in his

view, are incompatible with historical materialism. Brown thinks that the lesson Marxism

can learn from CR is the need to articulate the concepts of Marxian theory at the level of

generality of philosophy. More precisely, he suggests that a Marxist philosophy should

embrace some of the fundamental concepts of CR, such as structural causality, the

distinction between thought and mind-independent object, the notion of tendencies, the

relations between structure and agency. These notions, however, are stressed also within

other philosophical perspectives. Thus, Brown sees no particular reason for Marxism to

embrace a precisely CR philosophy and defends instead a materialist dialectics, as an

alternative Marxist philosophical position able to embrace and transcend CR.

It is not necessary here to go further into the details of the general coherence of Marxism

and CR. For my purpose, it is sufficient to re-establish the distinguishing features of the

capitalist mode of production along with its Marxian analysis and interpret them in CR

terms. This is sufficient to determine the constraining structure and the constraining

mechanisms of capitalism.1

Capitalism, in Marxs conception, is a system of generalised commodity production

where goods are produced, not for the direct consumption of producers, but for sale.

Marxs Capital(1867, 1885, 1894) does not begin with the isolated individual, but with the

commodity. The commodity contains all the distinguishing social relations of the capitalist

mode of production. Material things are not in their nature commodities. They becomecommodities only under particular historical conditions, with particular social relations.

The category of the commodity itself encompasses the categories of private property and

the market. Private property is a social relation that defines a series of duties and rights in

society and their distribution among agents. The existence of private property thus implies

the existence of power relations in the form of PTA.2 Through the market, then,

commodities receive a social appraisal (the price system) and, on this basis, are exchanged.

The exchange of commodities modifies the social system of duties and rights, i.e., modifies

the system of individual PTAs. The coexistence of private property and the market gives

rise to a specific form of PTA in the economy, namely, purchasing power: the power to buy.

In barter, the parties exchange their respective PTAs over commodities directly. In moneyexchange, by contrast, one party alienates a concrete PTA (the power to dispose of

a particular commodity) and receives money, i.e., the right to obtain commodities in the

future. Money can thus be seen as an abstract PTA over commodities.

1 In a review of Lawsons 2003 book Reorienting Economics, Pinkstone (2003) criticises the author fordefining economics in terms of social relations, but without any specific reference to power. He argues thatneoclassical economists do not deny that capitalists and workers are engaged in a social relationship, butmaintain that this relationship does not necessarily involve power. Pinkstone suggests that Marx is theeconomist who made social power the central focus of his explanation of production and distribution incapitalism. What follows can be seen as an attempt to follow Lawsons reorientation of economics by takingaccount of Pinkstones criticism.

2 Outside Marxism, the economic coercion emanating from property is investigated in particular by

institutionalist authors, such as Veblen (1934), Samuels (1979, 1984, 1994) and Schmid (1987). Therelations between Marxism and institutionalism in the analysis of power and classes are discussed by Klein(1992) and Dugger and Sherman (1994).


7/30/2019 2007-cje[1]

12/23

Private property and the market, however, are not sufficient to characterise capitalism.

The specificity of capitalism is that the process of commodification includes labour power

(the ability to work) as well. In capitalism, property is characterised not only by its legal

form as private property, but also by its unequal distribution between the two classes of

capitalists and wage workers. Property relations are thus also class relations in capitalism(Campbell, 1993).1 Marx discusses at length the conditions under which labour power

becomes a commodity. He explains that, for the wagelabour relation to arise, the worker

must be free in a double sense, that as a free man he can dispose of his labour-power as his

own commodity, and that on the other hand he has no other commodity for sale (Marx

1867, ch. 6). Marx then discusses the historical circumstances that produced on the one

side owners of money or commodities, and on the other men possessing nothing but their

own labour-power, and the mechanisms that reproduce these classes of people.2

Given the central role of private property and the market in capitalism, it is by the

analysis of purchasing power, as a form of power emanating from them, that we must start

our investigation of power relations in capitalism. In doing so, our focus must remain on

the particular transaction that distinguishes capitalism, namely the market exchange of

labour power.

5.1 Purchasing power

Purchasing power exists in every system based on private property and the market. It is

determined by the distribution of property rights and the array of prices, which define the

budget constraints of each agent. In general terms, the decision-making set of an agent

defines what the agent can and cannot do; in market relations, purchasing power defines

what the agent can and cannot buy.

The quantitative differences between agents purchasing power are a measure of the

existing asymmetries of PTA in the economic sphere.3 In capitalism, these quantitative

asymmetries produce an essential qualitative difference: on one side, there are people that,

given their (lack of) purchasing power, must selltheir labour power; on the other side, other

people, thanks to their purchasing power, can buy this labour power and make a profit from

1 Mattick (1993) maintains that Capital begins with the commodity only because Marx intended tocriticise classical political economy (which often starts with the commodity), but the real theoretical startingpoint is class relations. This interpretation contrasts the logical-historical interpretation of Marxs categoriessuggested By Engels (1886) and developed by Meek (1976), according to which the structure of Capitalreflects an idealised periodisation of history from pre-capitalist simple commodity production to propercapitalist production. Earlier critiques of this interpretation are developed by Banaji (1979) and Smith(1990). For an analysis of the significance of the opening chapters of Capital, see Pilling (1980, ch. 4).

2

Marx dedicates the whole Part VIII ofCapital 1, to study the process of primitive accumulation and thedivorce of the worker from his/her means of production and harshly criticises the authors that he calls vulgareconomists for their abstract speculations on the origin of capitalism. Unfortunately the same speculativemethod, based on hypothetical spontaneous interactions between isolated individuals, has today become thedistinguishing feature of new institutionalism. Although closer to a radical tradition, Putterman (1995)follows the same idea that hierarchical relations within the capitalist firm are the result of voluntary decisions.Similarly, Stiglitz (1975) contends that these relations were voluntarily selected by workers for efficiencyreasons. A Marxist reply to this position is offered by Braverman (1974), Marglin (1974, 1975), R. Edwards(1979) and P. K. Edwards (1990), who discuss the authoritarian content of the work relation and thepractices of extraction of the maximum possible amount of labour from a given quantity of labour power.More generally, historians of labour mostly agree that capitalist hierarchies were imposed with ampleresistance from workers and independent producers. Within Marxism, this process is analysed by Thompson(1978, 1993), Hobsbawn (1964), Marglin (1974, 1975, 1991) and Rule (1986); Landes (1969), Berg (1984,1991) and Pollard (1965) reach the same conclusion from a technology oriented perspective.

3

Within the institutionalist tradition, Schutz (1995) discusses the asymmetric distribution of purchasingpower as a potential cause of relations of POS in the market. The author, however, does not develop anyanalysis of this asymmetry as an essential aspect of the capitalist economic structure.

12 of 23 G. Palermo

7/30/2019 2007-cje[1]

13/23

it. Without mechanisms capable of reproducing an asymmetrical distribution of con-

straints on (classes of) individuals, there would be no stable labour market, and wage

labour might not be the dominant form of production. This entails a form of economic

compulsion, mostly indirect, operating through economic constraints, which brings a class

of persons to work (voluntarily) for another class of persons. This form of compulsion isnot the consequence of deliberate choices of particularly powerful individuals, but rather

the effect of the impersonal mechanisms that govern market prices.

5.2 Market power and authority

POS takes different forms in production and in circulation: in production, it takes the form

of authority relations; in circulation, that of market power.

Although the worker and the capitalist meet in the market on the basis of equal right (but

with asymmetric purchasing power), the fact that one sells labour power and the other buys

it produces a direct relation of POS in the process of production. As we have seen, the

champions of the liberal view, Alchian and Demsetz (1972), deny the existence of anyasymmetry between the worker and the capitalist on the ground that their interpersonal

relation is not different from any other market relation. Workers and capitalists, according

to them, are generic individuals, not social figures. As Marx explained 120 years before,

however:

The worker leaves the capitalist, to whom he has sold himself, as often as he chooses, and thecapitalist discharges him as often as he sees fit . . . But the worker . . . cannot leave the whole classof buyers, i.e., the capitalist class, unless he gives up his own existence. He does not belong to thisor that capitalist, but to the capitalist class; and it is for him to find his man i.e., to find a buyerin this capitalist class. (Marx 1847A, ch. 2)

When a grocer sells a kilo of coffee, he/she alienates a piece of his/her property and transfersto somebody else his/her power to do with the coffee as he/she likes. When a worker sells

his/her labour power, he/she alienates a piece of his/her life and transfers to somebody else

his/her power to do with himself/herself as he/she likes. The exchange of labour power

involves interpersonal relations directly. The PTA that the capitalist obtains in exchange

for his/her money is itself a POS. He/she gives purchasing power to the worker and acquires

authority over him/her.

Clearly, the authority relationship is never absolute, since, as for any commodity, the

utilisation of labour power is regulated by law and institutional rules. However, within the

existing norms, the POS involved in the authority relationship is absolute, exactly as

absolute as the PTA conferred by the ownership of any other commodity.In the sphere of circulation, an agent can influence the action of others only by having

some control over the price system. Market power is thus the form of POS in circulation. It

is defined as the power of the seller [buyer] to fix the price above [below] the price fixed by

other sellers [buyers]. This form of economic power is not specific to capitalism; it is rather

a potential feature of all systems based on private property and the market, for its existence

presupposes market exchange, but not necessarily the market exchange of labour power.

Both these forms of economic POS presuppose purchasing power as a form of economic

PTA. In order to have authority in production or market power in circulation, one must

first have purchasing power, for the obvious reason that, without the power to buy, one

clearly cannot buy labour power, or any other commodity, at whatever price. The problem

now is to explain how purchasing power on the one hand, and authority and market poweron the other, are continuously transformed into one another and reproduced.


7/30/2019 2007-cje[1]

14/23

5.3 The class structure of capitalism

The constraining structure of capitalism is characterised by class relations. In Marxs work,

although the history of mankind is a history of class struggle (Marx and Engels, 1848),

there is no systematic discussion of how to define social classes.1 The analysis of the

capitalist mode of production in its pure form, developed in Capital, is based on a classstructure with two main classes: capitalists and workers. In the study of concrete social

formations (Marx, 1847B, 1852; Marx and Engels, 1848), however, Marx does not

hesitate to consider a greater number of classes. This has prompted different interpreta-

tions of the theoretical status of classes in Marxs theory. According to an economistic

interpretation, in Capital, classes are defined solely in terms of the economic structure, by

the relation of each individual to the means of production whilst, in the analysis of concrete

social formations, classes must be defined also in terms of political and ideological

structures. This interpretation, however, has been criticised for its unjustified separation of

the economic structure, on the one hand, and political and ideological structures, on the

other, as if the capitalist mode of production could effectively be defined in purely

economic terms. Poulantzas (1973, ch. 2), in particular, argues that classes are the effect of

the whole set of social structures that characterise the capitalist mode of production, and

not solely of the economic structure.2

Here we do not need to develop the complex relations between the economic, the

political and the ideological structures of capitalism. The economic separation of the

population into buyers and sellers of labour power is sufficient to determine an essential

aspect of class relations in capitalism. This is not to reduce social classes to an asymmetry

in economic constraints; it is rather to recognise that the asymmetric distribution of

purchasing power is an essential aspect of every definition of the class structure of

capitalism.

With these caveats, capitalist class structure can be characterised by capitalistsmonopoly of the means of production. Abstractly, private property and the market might

exist without class divisions. Class monopoly of the means of production, however, is not

a historical accident. It is rather a necessary condition for the development of wage labour

and for the functioning of capitalism. Without such an asymmetry between buyers and

sellers of labour power, no stable wage labour relation would exist. Considered from the

viewpoint of constraints, wage labour exists only as long as a class of people is constrainedto

sell labour power. The other side of the coin is that another class of people is enabledto buy

labour power. This is the essential asymmetry of the capitalist constraining structure

(which capitalists might well call its enabling structure): workers are constrained by their

lack of purchasing power; capitalists are enabled by their possession of it.

5.4 Competition

The main mechanism that regulates the reproduction of class relations in capitalism is

competition. Competition is generally conceived of as a state of affairs (neoclassical

1 In Capital 3, there is a draft chapter dedicated to the subject. However, the chapter is incomplete, and itspurpose seems to be more to point out the problematic aspects of defining classes than to provide theblueprints for such a project.

2 Cf., also Poulantzas (1986). Colletti (1972) criticises Kautskys and Plekhanovs orthodoxy andBernsteins revisionism for their separation of the economic structure from the other social structures ofcapitalism, and argues that Marxs economic sphere in fact embraces the production of things as well asideas, and material production as well as the production of social relations. In developing Collettis

interpretation, Sayer (1987) maintains that the attempt to separate forces and relations of production, or baseand superstructure, as if they referred to different bits of empirical reality, presupposes an atomistic ontologythat is extraneous to Marxs.

14 of 23 G. Palermo

7/30/2019 2007-cje[1]

15/23

competition, with large numbers and small dimensions) or as a process (Austrian

conception of the market as a discovery procedure). However, as in the case of power,

these conceptions only confuse the mechanisms that govern actual-empirical phenomena

with the phenomena themselves. I analyse competition, using CR ontological insights, as

a causal mechanism. This mechanism is not specific to capitalism, but exists in othermodes of production as well. In capitalism, however, competition takes a specific form,

which depends on the specific social relations of such a mode of production. As Marx

(1867, ch. 12) put it, a scientific analysis of competition is not possible, before we have

a conception of the inner nature of capital, just as the apparent motions of the heavenly

bodies are not intelligible to any but him, who is acquainted with their real motions,

motions which are not directly perceptible by the senses.

According to Marx and Engels, the development of economic competition is to a large

extent a consequence of private property. In Engels words,

because private property isolates everyone in his own crude solitariness, and because, neverthe-

less, everyone has the same interest as his neighbour, one landowner stands antagonisticallyconfronted by another, one capitalist by another, one worker by another. In this discord ofidentical interests resulting precisely from this identity is consummated the immorality ofmankinds condition hitherto; and this consummation is competition. (Engels, 1844)

In turn, competition is also the cause of the tendency for private property relations to

expand, both extensively (geographical expansion of market relations) and intensively

(commodification of all aspects of nature and human life). This means that capitalism is by

its nature dynamic and self-expanding, and that what we today call globalisation is in fact

an intrinsic aspect of capitalist development. Competition and private property are thus

linked by a dialectical relation, which characterises the working and development of the

capitalist mode of production.In Marxs and Engels conception, unlike mainstream economics, competition is not the

result of the opposition of private interests of atomistic individuals. Human interests, with

a highly developed division of labour, are necessarily social interests. Competition,

therefore, is not merely a mechanism whereby isolated individuals interact, but a mode

of coordination ofsocial individuals, with convergent and divergent social interestssocial

interests that, in capitalism, depend in an essential way on class relations.

The social nature of competition in Marxs and Engelss critique has been the object of

debate within Marxism. In fact, in their analysis of competition, the two authors pursue

slightly different theoretical objectives. Engels intends to show that all the categories of

political economy and the realities to which they correspond presuppose both competitionand private property. In his critique, however, Engels explains competition as a conse-

quence of private property, but does not really explain private property. It is Marx that

undertakes the task of explaining the origins of private property as well (Clarke, 1991).

In fact, particularly in his early works, Marx focuses on the relations of private property to

another social relation, namely alienated labour, which suggests a more articulated relation

between private property and competition. As Marx explains, in order for labour to be

appropriated in the form of property, it must first take the form of alienated labour. In this

sense, although private property appears as the basis and cause of alienated labour, it is in

fact its consequence . . . Later, however, this relationship becomes reciprocal (Marx 1844).

To put it differently, Marx arrives at his social conception of competition via its relations

with alienated labour; Engels focuses instead on the necessarily asymmetrical distributionof private property between social classes. But commodification of labour power is for both


7/30/2019 2007-cje[1]

16/23

of them the theoretical and historical reason for the class dimension of competition in

capitalism.1 Let us thus consider how competition regulates the reproduction of class

relations.

Engels (1845, ch. 3) defines competition [as] the completest expression of the battle of

all against alla battle fought not between the different classes of society only, but alsobetween the individual members of these classes (cf., also Engels, 1884).2 Competition

between classes and that within classes are dialectically linked. Consider first the struggle

between classes. In this struggle, like in any other, the stronger often wins. And it is not

difficult to understand that workers and capitalists do not have the same strength, for the

former must work to live, whilst the latter can live on their capital. The asymmetric

strength between classes influences competition within classes as well: competition tends to

be stronger within the weaker class, i.e., on the workers side. This asymmetry, in turn,

reinforces the asymmetric strength between classes, since the stronger the competition

within one class, the weaker the whole class in its relation to the other.

This asymmetric competition between and within classes regulates wage dynamics and

unemployment or, in Marxs words, the industrial reserve army of labour (Marx 1867,

ch. 22). The industrial reserve army is an essential aspect of capitalismalthough today in

many advanced economies it takes the form of a reserve army of the underemployed, that

is, an army of flexible workers whose tasks and working-hours are adjusted to suit the

demands of productive conditions. First, it exerts pressure on the wages and working

conditions of employed workers. Second, in periods of expansion, it cushions the effects

that an increasing demand for labour may have on the labour market. But unemployment

is also a consequence of capitalist accumulation. Very simply, the surplus population could

all be put to work if the length of the working day were reducedsomething manifesting

itself in the recent French workers drive for a 35-hour week. The point, however, is that

there is no automatic tendency for this to occur, and capitalists surely do not have aninterest in thiswhich is why, in the French case, they vehemently opposed the law limiting

the working week to 35 hours, until it was finally abolished. More generally, this explains

the historical hostility of the bourgeoisie towards workers associationsa hostility whose

concrete expression ranges from legal prohibition in the past to ideological condemnation,

as a violation of perfect competition, in more recent times. As Engels (1845) wrote,

competition of the workers among themselves is . . . the sharpest weapon against the

proletariat in the hands of the bourgeoisie. But all this is secondary; the point is rather that

competition operates on the capitalists side as well, forcing them to extract as much labour

as possible from active workersa condition which is necessary for minimising costs and

remaining in business.Consider now wage determination. When competition among workers is strong, the

wage reaches the subsistence level, a level below which there is no reason to work. This

limit is relative and socially determined by the average needs and the grade of civilisation

of the workers (Engels 1845). Symmetrically, the maximum wage is determined by the

competition of capitalists among themselves. If workers need capitalists (to live), capitalists

need workers as well (to make profits). Competition among capitalists for employing new

workers tends to increase when the demand for the good they produce increases and when

1 Starting from the recognition of Marxs and Engels slightly different theoretical objectives, Clarke(1994) discusses the different implications of their analysis of competition in the study of crisis in capitalism.

2

This class dimension of competition is completely neglected by the individualist approach, whichassumes that competition is a consequence of human nature. Human nature and the market, in turn, are notseen as products of history but as everlasting realities.

16 of 23 G. Palermo

7/30/2019 2007-cje[1]

17/23

the reserve army is small. In these conditions, each capitalist prefers to pay slightly higher

wages rather than miss out on any profits. The wage tends thus to increase above the

subsistence level. However, as soon as this risks reducing the ordinary average profit,

capitalist competition loses momentum, and the wage no longer increases.

These processes must not be understood as empirical descriptions, but rather asexplanations of the tendencies produced by the interplay of competition within and

between classes. In modern capitalist economies, wage determination is a complex process

involving a whole web of social structures and institutions that can inhibit or reinforce these

tendencies. But even when these tendencies do not show themselves as empirical

regularities, they still operate as forces caused by the capitalist mode of social interaction.

To these economic tendencies produced by competition, Marx and Engels (1848)

respond politically: Working men of all countries unite! Association is in many respects

the opposite of competition. Unlike competition, however, this mechanism is not a direct

consequence of the economic structure of capitalism, and its concrete working depends on

class-consciousness and political action. More importantly, its existence is itself condi-

tioned by the asymmetric power relationship between classes (Marx, 1867; Engels, 1844,

1845, 1847; Marx and Engels, 1848). It is not historically accidental that associations

emerged quite spontaneously (and legally) among capitalists, whilst they encountered

many obstacles among workers and had to remain secret for quite some time. In fact, the

right to associate is itself a political victory for the workers movement, a result of class

struggle. In our analysis of the essential power relations of capitalism, the difference with

competition is that association is not necessary to the reproduction of the capitalist mode of

production (indeed the Wal-Mart model of capitalism without trade unions is not only

possible, but also very profitable). This is why Marx and Engels discuss the mechanism of

association mainly in their studies of concrete social formations and in their political works,

where association takes the feature of a conscious subjective response to the objectivecoercive law of competition. However, in the analysis of the capitalist mode of production

at a high level of abstraction, competition alone is the necessary, the essential, mechanism

that governs and reproduces class relations.

5.5 Capitalism as a system of power

The class structure of capitalism and the mechanism of competition make capitalism

a system of power. Within this system, PTA and POS are constantly reproduced and

transformed into one another. Consider first the transformation and reproduction of

purchasing power and authority. At time t, under the pressure of competition and becauseof the asymmetric distribution of purchasing power in society, K buys Ls labour power

(i.e., he/she exercises his/her purchasing power) and acquires authority (the right to

command) over L for the period (t, t 1). During this period, by virtue of his/her authority,

K extracts as much labour as possible from Ls labour power. At time t 1, K sells the

product and recovers the purchasing power he/she had anticipated (and probably more); at

the same time, L spends the purchasing power gained with the sale of his/her labour power

without, in general, accumulating enough to become independent of wage labour. This

process is relatively stable and tends to reproduce the initial class division. There is no need

for capitalists and workers to confront each other as classes for class relations to be

reproduced. The power relationship of capitalists over workers exists independently of

class-consciousness, and the tendencies it produces are independent of the wishes of bothindividual capitalists and workers and cannot be reduced to the choice of any of them.


7/30/2019 2007-cje[1]

18/23

Consider now the process of transformation and reproduction of purchasing power and

market power. This process is characterised by a tendency towards the concentration and

centralisation of capital (Marx 1867, ch. 25). In the process of competition, large firms use

more advanced technologies, exploit better increasing returns to scale and the division of

labour. This allows them to obtain higher profits and to accumulate faster. The result isa tendency towards concentration of the social means of production. This process is further

reinforced by the tendency to mergers and take-overs that proceed with capitalist

accumulation. In analysing this process, Marx points out the active role played by the

credit system and the acceleration prompted by crisis. The effect of these processes is that

fewer and fewer capitalists accumulate more and more capital. Competition tends thus to

produce monopoly, i.e., market power. The exercise of market power, in turn, reinforces

the asymmetric distribution of purchasing power.1

These tendencies, let me repeat, are not empirical regularities: first, there exist other

tendencies and counter-tendencies that govern empirical reality; second, social interaction

depends also on subjective factors, such as class-consciousness, the ability to associate, the

structure of political power, and much else. These tendencies are rather structural

necessitiesthe consequences of social relations that are necessary for the working and

reproduction of the system.

6. Conclusions

Economic power takes different forms in capitalism. Mainstream economics, however,

considers only the surface of interpersonal relations, in which power relations manifest

themselves in the form of POS. This restriction occurs without any systematic analysis of

the relations between POS and PTA and without any investigation of the factors thatgovern the evolution of the distribution of PTA in society. The cause of this, I have argued,

lies in the ER ontology that underlies mainstream economics.

In this paper, I have drawn upon CR to develop a sophisticated ontology of power

relations under capitalism, an ontology in which empirical and observable forms of power

are put in relation to the structures and mechanisms that govern them. In line with Marxs

method and analysis, I have shown that capitalism is a system of power, in which the forms

of POS (market power and authority) depend on the distribution of PTA (purchasing

power) in society. These phenomenal forms of power are governed by the constraining

structure (characterised by class relations) and the constraining mechanism that

reproduces it (competition).

This ontological investigation leads us to reject the mystifying conception of mainstream

economics of capitalism as being possibly free from power relations. The problem is not

simply that neoclassical perfect competition does not exist in reality, but rather that such

a hypothetical system could not work and reproduce itself without forms of impersonal

coercion produced by structures and mechanisms that are not necessarily detectable

empirically. The ontology of mainstream economics cannot consider these forms of social

coercion because, more generally, it does not allow the distinction between appearances

and essence, between formal and substantial power relations. If capitalist economic

relations are power relations, however, it is not because they necessarily appear as such

empirically, but because capitalism is in its essence a system of power.

1 Concentration and centralisation also reinforce the reproduction of the reserve army by creating theconditions for economising and rationalising labour.

18 of 23 G. Palermo

7/30/2019 2007-cje[1]

19/23

Bibliography

Alchian, A. A. and Demsetz, H. 1972. Production, information costs and economicorganization, American Economic Review, vol. 62, no. 1, 77795

Alchian, A. A. and Woodward, S. 1987. Reflections on the theory of the firm, Journal of

Institutional and Theoretical Economics, vol. 143, no. 1, 11036Ankarloo, D. and Palermo, G. 2004. Anti-Williamson. A Marxian critique of new institutionaleconomics, Cambridge Journal of Economics, vol. 28, no. 3, 41329

Archer, M. 1995. Realist Social Theory: The Morphogenetic Approach, Cambridge, CambridgeUniversity Press

Archer, M., Bhaskar, R., Collier, A., Lawson, T. and Norrie, A. (eds) 1998. Critical Realism:Essential Readings, London, Routledge

Arendt, H. 1970. On Violence, London, The Penguin PressBachrach, P. and Baratz, M. S. 1970. Power and Poverty. Theory and Practice, New York, Oxford

University PressBanaji, J. 1979. From the commodity to capital: Hegels dialectic in Marxs Capital, in Elson, D.

(ed.), Value: The Representation of Labor in Capitalism, London, CSE BooksBartlett, R. 1989. Economics and Power: An Inquiry into Human Relations and Markets,

Cambridge, Cambridge University PressBerg, M. 1984. The Age of Manufactures, London, FontanaBerg, M. 1991. On the origins of capitalist hierarchy, in Gustavsson, B. (ed.), Power and Economic

Institutions, Aldershot, Edward ElgarBhaskar, R. 1978. A Realist Theory of Science, Hemel Hempstead, Harvester PressBhaskar, R. 1979. The Possibility of Naturalism, Hemel Hempstead, Harvester PressBhaskar, R. 1986. Scientific Realism and Human Emancipation, London, VersoBhaskar, R. 1993. Dialectic: The Pulse of Freedom, London, VersoBowles, S. 1985. The production process in a competitive economy: Walrasian, Marxian and

Neo-Hobbesian models, American Economic Review, vol. 75, 1636Bowles, S. and Gintis, H. 1988. Contested exchange: political economy and modern economic

theory, American Economic Review, vol. 78, 14550

Bowles, S. and Gintis, H. 1993A, Power and wealth in a competitive capitalist economy,Philosophy and Public Affairs, vol. 21, no. 4, 32453

Bowles, S. and Gintis, H. 1993B. The revenge of homo economicus: contested exchange and therevival of political economy, Journal of Economic Perspectives, vol. 7, no. 1, 83102

Bowles, S. and Gintis, H. 1993C. Post Walrasian political economy, in Bowles, S., Gintis, H. andGustafsson, B. (eds), Markets and Democracy: Participation, Accountability and Efficiency,Cambridge, Cambridge University Press

Bowles, S. and Gintis, H. 1993D. The democratic firm: an agency-theoretic evaluation, inBowles, S., Gintis, H. and Gustafsson, B. (eds), Markets and Democracy: Participation,Accountability and Efficiency, Cambridge, Cambridge University Press

Bowles, S. and Gintis, H. 1994. Power in economic theory, in Arestis, P. and Sawyer, M. (eds),The Elgar Companion to Radical Political Economy, Aldershot, Edward Elgar

Braverman, H. 1974. Labor and Monopoly Capital, New York, Monthly Review PressBrown, A., Fleetwood, S. and Roberts, J. M. (eds) 2001. Critical Realism and Marxism, London,Routledge

Brown, A., Fleetwood, S. and Roberts, J. M. 2001. The marriage of critical realism andMarxism: happy, unhappy or on the rocks?, in Brown, A., Fleetwood, S. and Roberts, J. M.(eds), Critical Realism and Marxism, London, Routledge

Campbell, M. 1993. Marxs concept of economic relations and the method of capital, inMoseley, F. (ed.), Marxs Method in Capital. A Reexamination. Atlantic Highlands, HumanitiesPress

Cheung, S. N. S. 1983. The contractual nature of the firm,Journal of Law and Economics, vol. 26,no. 1, 121

Cheung, S. N. S. 1987A. Economic organization and transaction costs, in Eatwell, J., Millgate,M. and Newman, P. (eds), The New Palgrave: a Dictionary of Economics, London: Macmillan

Cheung, S. N. S. 1987B. Common property rights, in Eatwell, J., Millgate, M. and Newman, P.(eds), The New Palgrave: a Dictionary of Economics, London, Macmillan


7/30/2019 2007-cje[1]

20/23

Cheung, S. N. S. 1992. On the new institutional economics, in Werin, L. and Wijkander, H.(eds), Contract Economics, Oxford, Basil Blackwell

Clarke, S. 1991. Marx, Marginalism and Modern Sociology: From Adam Smith to Max Weber, 2ndedn, London, Macmillan

Clarke, S. 1994. Marxs Theory of Crisis, London, Macmillan

Coase, R. H. 1937. The nature of the firm, Economica, vol. 4, 386405Colletti, L. 1972. Bernstein and the Marxism of the second international, in Colletti, L. (ed.),

From Rousseau to Lenin, London, New Left BooksCollier, A. 1979. Materialism and explanation in the human sciences, in Mepham, J. and Ruben,

D. H. (eds), Issues in Marxist Philosophy, Brighton, The Harvester PressCollier, A. 1989. Scientific Realism and Socialist Thought, Hemel Hempstead, Harvester

WheatsheafCollier, A. 1994. Critical Realism: An Introduction to Roy Bhaskars Philosophy, London, VersoCousins, M. and Hussain, A. 1984. Michel Foucault, Basingstoke, MacmillanCreaven, S. 2001. Marxism and Realism: A Materialistic Application of Realism in the Social

Sicences, London, RoutledgeCreaven, S. 2003. Marx and Bhaskar on the dialectics of freedom, Journal of Critical Realism,

vol. 2, no. 1, 6393Dahl, R. A. 1961. Who Governs? Democracy and Power in an American City, New Haven, YaleUniversity Press

Dow, G. K. and Putterman, L. 1999. Why capital (usually) hires labor: an assessment ofproposed explanations, in Blair, M. M. and Roe, M. (eds), Employees and CorporateGovernance, Washington D.C., Brookings Institution Press

Dow, G. K. and Putterman, L. 2000. Why capital suppliers (usually) hire workers: what we knowand what we need to know, Journal of Economic Behavior and Organization, vol. 43, 31936

Dugger, W. M. 1984. An Alternative to Economic Retrenchment, New York, Petrocelli BooksDugger, W. M. 1989. Corporate Power, New York, Greenwood PressDugger, W. M., and Sherman, H. J. 1994. Comparison of Marxism and institutionalism, Journal

of Economic Issues vol. 28, no. 1, 10127Edwards, P. K. 1990. The politics of conflict and consent: how the labor contract really works,

Journal of Economic Behavior and Organization, vol. 13, 4164Edwards, R. 1979. Contested Terrain, New York, Basis BooksEhrbar, H. G. 1998. Marxism and Critical Realism, http://www.econ.utah.edu/ehrbar/marxre.pdfEhrbar, H. G. 2001. Critical realist argument in Marxs Capital, in Brown, A., Fleetwood, S. and

Roberts, J. M. (eds), Critical Realism and Marxism, London, RoutledgeEngels, F. 1844. Outline of a Critique of Political Economy, The Marx/Engels Internet Archive,

available at http://www.marxists.orgEngels, F. 1845. The Conditions of the Working-Class in England, The Marx/Engels Internet

Archive, available at http://www.marxists.orgEngels, F. 1847. The Principles of Communism, The Marx/Engels Internet Archive, available at

http://www.marxists.orgEngels, F. 1872. On Authority, The Marx/Engels Internet Archive, available at http://

www.marxists.orgEngels, F. 1878. Anti-Duhring. Herr Eugen Duhrings Revolution in Science, The Marx/Engels

Internet Archive, available at http://www.marxists.orgEngels, F. 1884. The Origin of the Family, Private Property and the State, The Marx/Engels Internet

Archive, available at http://www.marxists.orgEngels, F. 1886. Ludwig Feuerbach and the End of Classical German Philosophy, The Marx/Engels

Internet Archive, available at available at http://www.marxists.orgFleetwood, S. (ed.) 1999. Critical Realism in Economics: Development and Debate, London,

RoutledgeFleetwood, S. 2001. What kind of theory is Marxs labour theory of value? A critical realist

inquiry, Capital & Class, no. 73, 4177Foucault, M. 1977. Discipline and Punish: The Birth of the Prison, London, Allen LaneFoucault, M. 1980A. The History of Sexuality, Volume 1: An Introduction, London, PenguinFoucault, M. 1980B. Power/Knowledge: Selected Interviews and Other Writings by Michel Foucault

19721977, edited by C. Gordon, New York, Pantheon

20 of 23 G. Palermo

7/30/2019 2007-cje[1]

21/23

Foucault, M. 1990. Politics, Philosophy, Culture: Interviews and Other Writings 19771984, NewYork, Routledge

Fourie, F. C. v. N. 1989. The nature of firms and markets: do transaction cost theories help?, TheSouth African Journal of Economics, vol. 57, no. 2, 14260

Fourie, F. C. v. N 1991. The nature of the market: a structural analysis, in Hodgson, G. M. and

Screpanti, E. (eds), Rethinking Economics. Markets, Technology and Economic Evolution,Aldershot, Edward Elgar

Fourie, F. C. v. N. 1993. In the beginning there were markets?, in Pitelis, C. (ed.), TransactionCosts, Markets and Hierarchies, Oxford, Basil Blackwell

Galbraith, J. K. 1983. The Anatomy of Power, Boston, Houghton MifflinGintis, H. 1976. The nature of labor exchange and the theory of capitalist production, Review of

Radical Political Economics, vol. 8, no. 2, 3654Gintis, H. 1989. Financial markets and the political structure of the enterprise, Journal of

Economic Behavior and Organization, vol. 11, 31122Gintis, H. and Ishikawa, T. 1981. Wages, work discipline, and unemployment, Journal of

Japanese and International Economics, vol. 1, 195228Goldberg, V. P. 1976. Toward an expanded economic theory of contract, Journal of Economic

Issues, vol. 10, no. 1, 4561Goldberg, V. P. 1980. Bridges over contested terrain: exploring the radical accountof the employment relationship, Journal of Economic Behavior and Organization, vol. 1,24974

Gunn, R. 1989. Marxism and philosophy: a critique of critical realism, Capital & Class, no. 37,86116

Hart, O. 1995. Firms, Contracts, and Financial Structure, Oxford, Oxford University PressHart, O. 1996. An economists view of authority, Rationality and Society, vol. 8, 37186Hart, O. and Moore, J. 1990. Property rights and the nature of the firm. Journal of Political

Economy, vol. 98, 111958Hobsbawn, E. J. 1964. Labouring Men: Studies in the History of Labor, London, Weidenfield &

NicolsonJensen, M. and Meckling, W. 1976. Theory of the firm: managerial behaviour, agency costs and

ownership structure, Journal of Financial Economics, vol. 3, 30560Joseph, J. 2001. Five ways in which critical realism can help Marxism, in Brown, A., Fleetwood,

S. and Roberts, J. M. (eds), Critical Realism and Marxism, London, RoutledgeKlein, N. 2000. No Logo, Toronto, Alfred A. KnopfKlein, P. A. 1992. Institutionalists, radical economists, and class, Journal of Economic Issues,

vol. 26, no. 2, 53544Landes, D. 1969. The Unbound Prometheus, Cambridge, Cambridge University PressLawson, T. 1997. Economics and Reality, London, RoutledgeLawson, T. 2003. Reorienting Economics, London, RoutledgeLukes, S. 1974. Power: A Radical View, London, MacmillanMagill, K. 1994. Against Critical Realism, Capital & Class, no. 54, 11336Marglin, S. A. 1974. What do bosses do? The origins and functions of hierarchy in capitalist

production, Part 1, Review of Radical Political Economics, vol. 6, no. 2, 60112Marglin, S. A. 1975. What do bosses do? The origins and functions of hierarchy in capitalist

production, Part 2, Review of Radical Political Economics, vol. 7, no. 1, 2037Marglin, S. A. 1991. Understanding capitalism: control versus efficiency, in Gustavsson, B.

(ed.), Power and Economic Institutions, Aldershot, Edward ElgarMarx, K. 1844. The Economic and Philosophic Manuscripts of 1844, The Marx/Engels Internet

Archive, available at http://www.marxists.orgMarx, K. 1847A, Wage Labour and Capital, The Marx/Engels Internet Archive, available at http://

www.marxists.orgMarx, K. 1847B, The Poverty of Philosophy, The Marx/Engels Internet Archive, available at http://

www.marxists.orgMarx, K. 1852, The 18th Brumaire of Louis Napoleon, The Marx/Engels Internet Archive, available

at http://www.marxists.orgMarx, K. 1859, A Contribution to the Critique of Political Economy, The Marx/Engels Internet

Archive, available at http://www.marxists.org

The ontology of ec

2007-cje[1]

Documents

Transcript of 2007-cje[1]